Comparative Effectiveness of Family DSMES and Standard DSMES Among Diverse Populations

Approximately 11% of the United States population has type 2 diabetes mellitus, including 12% of Arkansas's population. Diabetes self-management education and support (DSMES) is recommended for people with diabetes (PWD) to support necessary self-management practices and improve health outcomes. Systematic reviews have documented the efficacy of family models of DSMES. To our knowledge, there has been no randomized comparison of the effectiveness of a family model and standard model of DSMES that: 1) evaluates curricula that incorporate elements required for program accreditation/recognition; 2) is implemented among general/diverse populations using curricula that is not culturally adapted; and 3) evaluates outcomes in PWD and family members/support persons. This study addressed important gaps in knowledge by comparing a family-based model of DSMES (Family-DSMES) to standard (individual focused) DSMES (Standard-DSMES). Family-DSMES included PWD and their family members/support persons in educational sessions which explicitly addressed diabetes self-management within a family context, educating both patients and family members/support persons, family goal setting to support the PWD, understanding supportive and non-supportive behaviors, and family behavioral changes.

The overarching research question was "Compared to Standard-DSMES, does Family-DSMES yield improved outcomes among diverse patients with type 2 diabetes mellitus and their family members/support persons?"

The primary outcome was change in hemoglobin A1c (HbA1c) between baseline and post-intervention among PWD. Secondary outcomes for PWD included other biometric outcomes (body mass index, and blood pressure), behavioral outcomes (diabetes self-care behaviors and medication adherence), and psychosocial outcomes (diabetes management self-efficacy, diabetes-related distress, diabetes-related support, family involvement in diabetes management, and diabetes-related quality of life). We also examined the duration of treatment effects by comparing outcomes collected at 6 months post-intervention, 12 months post-intervention, and 18 months post-intervention using HbA1c data extracted from electronic medical records.

PWD were randomly assigned to either the Family-DSMES arm or the Standard-DSMES arm. Family-DSMES was delivered in group sessions via telehealth by a certified diabetes care and education specialist (CDCES) to patients and their family members. The intervention included 10 hours of education delivered over 10 weeks in one-hour sessions. Family members in the Family-DSMES arm took part in educational sessions and data collection. Standard-DSMES was delivered in group sessions via telehealth by a CDCES to patients only. The intervention included 10 hours of education delivered over 10 weeks in one-hour sessions. Family members took part in data collection but did not take part in educational sessions.

We recruited participants through eight primary care clinics across the state of Arkansas. Initial eligibility was determined through medical record abstraction. Patients who were deemed potentially eligible were sent a recruitment letter from their clinic and the Principal Investigator with an explanation of the study, followed up with a phone call from study staff or being approached in the clinic to discuss interest in the study. Potentially eligible and interested PWD were asked to identify a family member/support person to participate in the study with them, who were also contacted by study staff to assess interest and eligibility.